Luisa Doria (Presenter)
Imperial College London
Bio: I started my journey with a BSc in Biology in University of Porto. I did my MSc in Biochemistry with specialization in Bio-molecular Methods from University of Aveiro also in Portugal. As part of my MSc research project, I focused on mass spectrometry-based lipidomic analysis in breast cancer cell lines there in University of Aveiro, in the Mass Spectrometry group from Department of Chemistry. I have also been awarded a research fellowship in January 2012 at the Department of Chemistry from the same university to continue my learning on mass spectrometry methodology in lipidomic analysis, not just in phospholipids but also in other lipid classes. My interests in this technique together with the study of diseases make me apply to this PhD, where I am going to continue to do lipidomic analysis with new approaches inside the mass spectrometry field to understand the lipid distribution behi
Authorship: Luisa Doria (1), James McKenzie (1), Anna Mroz (1) , Abigail Speller (1), David Phelps (1), Kirill Velsekov (1), Sadaf GhaemMaghami (1), Zoltan Takats (1)
(1) Imperial College of London
| Short Abstract Ovarian cancer is the fifth most common cancer among women, mainly due to the poor and vague prognosis and diagnosis. DESI-MS is an excellent technique to characterize different cancer types. It provides detailed spatial information within the sample, providing the opportunity to investigate tumour biology from an entirely new perspective with accurate biochemical information about each tissue type. Using DESI coupled to a TQ for diagnosis makes this technique quicker and less expensive. Furthermore, it opens the possibility to use DESI for targeted imaging and quantification. |
Long Abstract
Ovarian cancer is the fifth most common and the most lethal cancer among women, largely related to late diagnosis. Current modalities for detecting ovarian cancer are primarily based on imaging and serological biomarkers (such as carbohydrate antigen-125). Once the presence of a mass has been confirmed, its malignant potential must be determined through exploratory laparotomy and subsequent biopsies. After surgery the “gold standard‟ for tissue biopsy and final diagnostics is histopathological assessment. There is an increasing demand for the improvement of precision, efficiency and lower costs in ovarian cancer diagnosis. Mass spectrometry (MS) is a powerful tool employed increasingly in the field of lipidomics, allowing the identification, characterisation, and quantitation of various lipid species. The wide use of non-targeted and targeted mass spectrometry approaches for phospholipids and sphingolipids in clinical studies in the last decade shows the increased relevance of understanding the role of these molecules in the metabolism of different diseases. Studies investigating breast cancer and colorectal cancer with desorption electrospray ionisation mass spectrometry (DESIMS) have already shown the potential of this technique to topographically localize molecular information and supplement conventional histological classification systems, giving us the opportunity to investigate tumour biology from an entirely new perspective. As such, in this study different epithelial ovarian carcinomas were analysed by DESI-MS and DESI-MS/MS to characterize and supplement cancer diagnostics by examining lipidomic changes.
Epithelial serous ovarian carcinoma, borderline tumours as well as normal ovary (a total of 30 samples) were collected. Samples were then cryosectioned and analysed by DESIMS in positive and negative ion mode using an Exactive orbitrap mass spectrometer. Their lipid profile was analysed in order to determine the targeted approach to perform. The DESI source was then coupled with a triple quadrupole (TQ) mass spectrometer and all samples were analysed performing parent scan and neutral loss acquisitions for a semi-targeted analysis. In addition all tissue sections analysed by DESI (MS and MS/MS) were also stained with haematoxylin and eosin (H&E), scanned and examined by a histopathologist in order to align the optical and mass spectrometric image for precise selection of region of interests(epithelial carcinomas and borderline tumours, the respectively surrounding stroma and healthy (“control”) stroma from normal ovary).
Multivariate statistical models were created using epithelial carcinoma tissues, borderline tumours and normal stroma from normal ovary. The m/z range used was 600 to 1000 Da, covering the range of phospholipids present in the cells. The resulting model was very strong with a clear separation of the different sample types and a leave-one- patient-out cross-validation performance of more than 90% for both ion modes. In addition, multivariate statistical analysis was performed within each individual sample to identify different tissue types based on the corresponding histological image and performing supervised analysis using maximum margin criteria (MMC). Leave-one-region-out crossvalidation results exceeded 98% accuracy. Multivariate statistical model was also created to evaluate the different tissue types using the entire dataset. Supervised MMC analysis clearly differentiated all tissue types with a leave one patient out crossvalidation performance of ~80%. The lipids found in the DESI spectra using Exactive orbitrap mass spectrometer were identified to further elucidate the ovarian cancer biochemistry. The predominant lipids found in negative ion mode were phosphatidylethanolamines (PE), phosphatidylserines (PS) and phosphatidylinositol (PI). In positive ion mode, the two most abundant classes were PEs and phosphatidylcholines (PC). Several lipid classes presented intensity changes in their profiles: PE, PI, phosphatidic acid (PA) and ceramides. To better understand the intensity changes observed for these different lipid classes, a semi-targeted method was developed, where parent scan and neutral loss analyses of the specific phospholipid head groups were analysed on a TQ coupled with a DESI source. With this method we can have a semi-quantitative method for specific lipids and lipid classes capable of characterizing not just different groups of samples but also different tissue types within the same samples. The classes of phospholipids analysed with this approach were PI and PA in negative ion mode and PE, PC, and ceramides in positive ion mode. Using this semi-targeted approach, a clear separation of the different sample and tissue types based on just a few individual phospholipid classes could be achieved. Using DESI coupled to a TQ for diagnosis makes this technique quicker and less expensive. Furthermore, it opens the possibility to use DESI for targeted imaging and quantification.
Overall, with DESI-MSI, we can accomplish robust tissue characterization and identification of cancer and tissue specific lipid biomarkers which could lead us to a new method in cancer diagnosis and prognosis at a histology level.
References & Acknowledgements:
| Description | Y/N | Source |
| Grants | no | |
| Salary | no | |
| Board Member | no | |
| Stock | no | |
| Expenses | no |
IP Royalty: no
| Planning to mention or discuss specific products or technology of the company(ies) listed above: | no |